2722
J. Ory. Chem., Vol. 36, No. 8, 1970
GORDON
was refluxed for 20 min, cooled, and decanted from the nickel which was washed with three 10-ml portions of ether. The combined ethereal extracts were dried over magnesium sulfate and concentrated. The oil was then taken up in 2 ml of hexane and chromatographed on alumina (5-10 g, Alcoa F-20) with hexane to remove alcohol and ketone (produced in the metalammonia reduction). The eluent was concentrated and analyzed by glpc on a 9-ft 20% Carbowax 20M on firebrick column a t 150". cis-3,3,5-Trimethylcyclohexyl Ethyl EtherW-cis-3,3,5-Tri(20 g) wm converted to its acetate by methylcyolohe~anol~~ treatment with acetic anhydride (50 ml) and pyridine (10 g) (4 hr heating on steam bath). The acetate (23 g, 89%) boiled a t 72" (3.5 mm), 12% 1.4392. Anal. Calcd for C11H2002: C, 71.69; H, 10.94. Found: C, 72.07; H, 11.19. Reduction was carried out by the method of Pettit and coworkersS2 starting with 4.92 g of the acetate in 120 ml of boron trifluoride etherate which was added to 2.28 g of LiAlHa in 60 ml of ether. Reaction was effected with cooling for 90 min followed by 1-hr reflux. After decomposition with 10% sodium hydroxide the ether layer was concentrated, chromatographed over Alcoa F-20 alumina with hexane and concentrated again to give upon distillation 2.5 g (5501,) of cis-3,3,5-trimethylcyclohexyl ethyl ether, bp 77-80' (20 mm). Repetition of the chromatography yielded pure material, bp 48-49' (3.5 mm), n*'D 1.4350. Anal. Calcd for C11HZnO: C, 77.58; H, 13.02. Found: C, 77.64; H, 13.05. trans-3,3,5-Trimethylcyclohexylethyl ether was synthesized analogously from the trans alcohol.s1 The acetate, bp 60-60.5' (2 mm), ~ O 1.4399, D was obtained in 82% yield. Anal. Calcd for C.I I R C~, 71.69: : H, 10.94. Found: C, . ~~O 72.04; H, 11.15. The ether was obtained in 68% yield, bp 74-75' (20 mm), n% 1.4338. (31) E . L. Eliel and H. Haubenstock, J . Org. Chem., 26, 3504 (1961). (32) G. R. Pettit, U. R. Ghatak, B. Green, T. R. Kasturi, and D. M. Piatak, ibid., 26, 1685 (1961).
Anal. Calcd for CllHnzO: C, 77.58; H, 13.02. Found: C, 77.86: H. 13.40. trans-4-~-Butylcyclohexyl propyl ether was obtained analogously from trans-4-t-butylcyclohexanol and propionic anhydride, followed by reduction. The ether boiled a t 112-114' (10 mm), 12% 1.4476. Anal. Calcd for CUHZ~O:C, 78.72; H, 13.21. Found: C, 79.13; H, 13.20. trans,trans-4-t-ButylcyclohexyloxyethylDisulfide.-trans-4-tButylcyclohexyloxyethyl mercaptan of 98% stereoisomeric purity (Table IY) (0.95 g, 4.5 mmol) in 8 ml of 5% NaOH solution was treated with 0.5 ml of 30% hydrogen peroxide. Excess peroxide was destroyed with solid sodium metabisulfite and the product was extracted three times with a 20-ml portion of ether. Drying and concentration yielded 0.92 g (97%) white solid, recrystallized from ethanol to give white plates, mp 51.5-52'. Anal. Calcd for CZ~H,,O~SZ: C, 66.92; H, 10.77; 8, 14.89. Found: C, 67.16; H, 10.76; S, 14.77. p-Chloroethyl benzyl sulfide was prepared from 40 g (0.25 mol) of P-hydroxyethyl benzyl sulfidela and 120 ml of concentrated hydrochloric acid by refluxing (with a gas trap) for 4 hr. The mixture was poured into 200 ml of water, the organic layer separated, and the aqueous layer was twice extracted with 100-ml portions of ether. The combined organic product was washed once with water and once with concentrated aqueous NaC1, dried, and concentrated to give after distillation 33 g (75%) of pchloroethyl benzyl sulfide, bp 91.5" (0.6 mm), n% 1.5682. Anal. Calcd for CDHdX3: C, 57.89; €I, 5.93; C1, 18.99. Found: C, 58.44; H, 6.00; C1, 18.86.
Registry No.--cis-3,3,5-Trimet hy Icyclohexyl ethy1 et her, 24691-15-4 ; cis-3,3,5-trimet hylcyclohexanol (aceethyl tate) , 24691-16-5; trans-3,3,5-trimethylcyclohexyl ether, 24691-17-6 ; trans-3,3, 5-trimethylcyclohexanol (acetate), 24691-18-7 ; trans-4-t-butylcyclohexyl propyl ether, 24691-19-8; trans1truns-4-t-butylcyc10hexy10xyethyl disulfide, 24691-20-1; p-chloroethyl benzyl sulfide, 4332-51-8.
Aromatic Nitration by Silver Nitrate Impregnated Silicic Acid in the Presence of Carbon Tetrachloride JOHNE. GORDON Department of Chemistry, Kent State University, Kent, Ohio 44940 Received September 93, 1969 Carbon tetrachloride reacts rapidly with excess silver nitrate impregnated silicic acid at room temperature to produce carbon dioxide and a nitrating agent. With carbon tetrachloride in excess, phosgene and the nitrating agent are formed. Aromatic compounds me nitrated by this system under column-chromatographic or batch conditions. A reaction sequence involving trichloromethyl nitrate, phosgene, nitryl chloride, carbonyl nitrate, and dinitrogen pentoxide is proposed.
While chromatographing trans-stilbene on AgNOasilicic acid, we attempted to elute with carbon tetrachloride. The result was complete destruction of the stilbene with formation of some 4,4'-dinitrostilbene. We have observed nitration of several other aromatic compounds under these conditions, but none of these reactions occurs in the absence of carbon tetrachloride. Carbon tetrachloride is generally considered inert to silver nitrate in homogeneous solution.3 Petrenko-Kritschenko and Opotsky4 reported 3% reaction with 0.2 N silver nitrate in 95% ethanol a t 90" in 12 (1) J. E . Gordon, J . Chromatogr., 48, 532 (1970). (2) E. Wenkert, D. J. Watts, and L. L. Davis [Chem. Commun., 1317 (1969)],have recently reported nitration of phenols by AgNOi-silicic acid alone. This appears t o depend upon the presence of the acidic OH group. (3) See, e . ~ , R. , L. Shriner, R. C. Fuson, and D. Y.Curtin, "The Systematic Identification of Organic Compounds," 4th ed, John Wiley & Sons, Inc., New York, N. Y.,1956,p 136 ff. (4).'I Petrenko-Kritschenko and V. Opotsky, Ber., 69, 2131 (1926).
hr; the products were not examined. We were thus interested to attempt identification of the reactions involved.
Results Static batch experiments with benzene and anisole are symmsrized in Table I. The basic experimental results drawn from there and the Experimental Section follow. (1) Neither benzene nor anisole is nitrated by AgNOa-silicic acid alone (expt 8, 18). Both are nitrated in high yield by AgNOrc-silicicacid in the presence of CC14 (expt 16, 17). (2) The nitration reaction involves specifically Ag+, NOa-, CC14, and the silicic acid surface (expt 6, 8, 9; ref 4). (3) The nitration reaction is independent of illumination and the presence of air (expt 3, 4, 11, 12). (4) Phosgene is formed in quantity. Two independent determinations
SILVER NITRATEIMPREGNATED SILICIC ACID
J. Org. Chem., Vol. 56, No. 8, 1970
2723
TABLE I NITRATION OF BENZENE A N D ANISOLE -Reaction Expt
1 2 3 4 5 6
7 8 9 10 11
CaHe, mmol
2.74 2.74 2.74 2.74 220 2.74 2.74
CeHaOCHs, mmol
CClr,
cc14 -k
AgNOs,
mmol
mmol
AgNOa
C
4.26 4.70d 4.6 4.6 4.39 4.72 4.71 4.49 5.92
2.8 2.8 2.8 2.8 3.5 11 22
C C
C C
0 C
C
0
2.74
C
2.74 2.74 2.74 2.74 2.74 2.74 2.68
C C
12 C 13 C 14 C 15 C 16 C 17 2.25 C 18 2.82 0 19 9.1 0.975 20 9.1 0.975 21 9.1 0.975 22 9.1 0.975 23 9.1 0,975 a Volume ratios of mixed solvents: Mean of two experiments. e Mean 20% ortho, 80% para.
time, hr, CCla
+
+ ArH
Yield, mmol of ArNOz/mmol of limiting reagent
2.8 2.8 2.8 2.8 3.5 11 22 24 24
0.042 0.060 O.OOSd 0.14 0.14 0.0 0.0 0.0 0.0 0.0
AgNOs
24
4.70* 24 24 AgNOa 0.41 =k 0.02' 4.79 AgNOa 0.48 48 48 5.04 48 48 AgNOa 0.51 2.63 91 91 0.51 4.02 AgNOa 0.47 97 97 5.01 AgNOa 0.52 90 90 6.12 144 144 CeH6 1.01 4.68 CeH6OCHa 0.71f 21 21 5.34 70 AgN03 0.0 70 6.49 1.2 58 cc14 0.38 5.86 CCll 0.25 1.0 18 5.83 CCla 0.71 5.0 18 5-92 CCla 0.20 18 10 12.8 CCla 1.328 18 5.0 cCl&eH~,, 1:2: CClr-CcHc, 1:l: CC14-CHCla. 1:l. Based on ea 1 or 2. Large excess. of three experiments. f Isomer composition: 21% ortho, 79% pari. g Isomer composition:
yielded 0.27 and 0.24 mol COC12/mol of AgN03 (CC14 in excess). Phosgene production is complete in 2 hr. (5) Dinitrogen tetroxide appears in the AgN0,silicic acid-CCle reaction mixture after a few hours in the absence, but not in the presence of aromatic substrate. (6) Mass spectrometric observation of the AgR03-silicic acid-CCle reaction mixture reveals only phosgene, CC4, and traces of COz when CC14 is in excess, but, when CC1, is in deficiency, COz is present in quantity. (7) No chlorobenzene could be detected in the AgNO3-silicic acid-CCl4-C~H~reaction product. (8) With AgXO3 as the limiting reagent, the ArT\;Oz yield levels off after 48 hr (room temperature) at a limiting ArXO2:AgNOaratio of 0.50 =k 0.02 (expt 2, 10, 11, 13-15). (9) The stoichiometry with respect to cc14 could not be accurately determined owing to the slowness of the AgN03-CC14 reaction when CC1, is limiting, so that decomposition of the nitrating agent becomes competitive. Also, inhibition of the AghiOaCCL reaction by ArH (see 12 below) was so severe with CCL in deficiency that reasonable yields could be obtained only by allowing the AgX03-CCli reaction to proceed for some hours, using a large excess of AgNOasilicic acid, then adding excess ArH. The ArXOz yield passes through a maximum as the CC14-AgNOa reaction time increases (expt 20-22) , and the maximum ArN02:CC14 ratio observed is 1.32 (expt 23). (10) The nitrating agent cannot be eluted from AgNOasilicic acid with CCI4. (11) The ortho/para ratio in the nitration of anisole is 0.25 with CC1, in excess or deficiency (expt 17, 23). (12) Aromatic substrates inhibit the CCla-AgNOa-silicic acid reaction (expt 5, expt 19 vs. 21; see Experimental Section.) (13) Substitution of alumina for silicic acid in expt 11 reduces
-I
I
nitrobenzene production from 0.48 to 0.02 mol/mol of AgKOa.
Discussion We believe that eq 1 and 2 give the simplest description of these reactions which is capable of rationalizing all of the observations. When the observed nitroCCl,
+ 2AgN03 + ArH = 2AgCl
CCh
+ 4AgN03 + 2ArH
=
4AgC1
+ COClt + ArNOz + HNO,
(1)
+ 2ArN02 + COZ + 2"03
(2)
*
benzene production in presence of excess CCI, (0.50 0.02 mol/mol of &NO,) is corrected for the slow nitration of benzene by the nitric acid formed according to eq 1 and 2, the net yield becomes 85-90% of the theoretical yield for these processes. The formation of 0.25 mol COClz/mol AgKOa implies parallel occurrence of eq 1 and 2, i e . , competition between CC1, and COClz for the available AgN03 such that some COC1, remains when the Agl\'03 is exhausted. When CCl, is the limiting reagent, eq 2 should apply alone. The difficulties in establishing the stoichiometry under these conditions have been described. Correcting the maximum ArNO2:CCl4ratio observed (1.32) for nitration by the nitric acid formed according to eq 2 gives a net value of 1.0. Since we know that the nitrating agent is decomposing with time and that side-reaction products are formed in the nitration step, the experimental results require a ArNOz:CC14 stoichiometry >1:1. The detailed course of reactions 1 and 2 has not been rigorously established. However, both can be ac-
2724 J. Org. Chem., Vol. 36, No. 8, 1970
GORDON
counted for by individual steps which are in accord with our observations and for which reasonable precedents are available. The sum of eq 3-8 is eq 2; eq 1 is the sum
+ AgN03 +CClaONOz + AgCl CC1,ONOz ----t COClz + NOzCl NOzCI f AgNOa +Nz06 + AgCl COClz + 2AgN03 2AgCl + CO(0NOz)i CO(0NOz)z --t COz + Nz06 2ArH + 2NzOs ---f 2 A r N 0 ~+ 2HN03 CCla
(3) (4)
(5) (6)
(7) (8)
of eq 3-5 and half of eq 8. The trichloromethyl nitrate of eq 3 is not known, but trichloromethyl perchlorate can be prepared from CC14 and solid AgC104 in the presence of AgCl, and it decomposes to p h o ~ g e n e . ~Trichloromethanol is unknown; it decomposes to phosgene and HCL6 Equation 5 is a known reaction.' Acyl chlorides react with AgNOa to give acyl nitrates in analogy with eq The acyl nitrates undergo facile thermal disproportionation to N205 and carboxylic anhydride in analogy with eq 7.8 Dinitrogen pentoxide is known to nitrate benzene instantly in cc14 solution a t 20";in nitration by benzoyl nitrate, N20Eia the actual nitrating agentng Dinitrogen pentoxide also decomposes thermally to NOZ-WZO~ O2in the gas phase and in solution with a half-life of a few hours a t room temperature.1° Anisole is nitrated by Ne05 (or by 2ArCOON02 (ArC0)20 -t N2059)predominantly in the ortho position, whereas the CC1~-AgN03-silicicacid reagent gives the predominance of para nitration characteristic of nitronium ion nitrati0ns.l' However, it has been shown that the second-order nitration by molecular Ne06 in CC14 is replaced in the presence of "03 by higher order processes catalyzed by " 0 3 and involving solvated nitronium ion as the nitrating agent.I2 There is physical evidence that drying a t 120-180" removes all of the adsorbed water from the silicic acid surface,13 but treatment of such material with dichlorodimethylsilane produces hydrolysis products of (CH&SiCl,, l 4 I n view of this and because the dried AgN03-silicic acid samples used in this work were not handled in a strictly anhydrous atmosphere, it is most likely that some hydrolysis of X Z Ooccurs; ~ the orientation pattern observed is therefore reasonable. The origin of the surface catalysis of eq 3 is of some interest. This is very likely catalysis by AgCl. The reaction of CC1, with solid AgC104 requires initiation by a small amount of HCL5 Also, it has recently been shown that the familiar reaction 9 proceeds simulta-
+
CzHd
+ AgNOs + HzO +CzHbOH + AgI + HNOa
(9)
(5) L. Birckenbach and J. Goubeau, Naturwissenschajten, 18, 530 (1930); Ber., 6 4 B , 218 (1931). (6) K. Elbs and K. Kratz, J . Prakt. Chem., 56 [ Z ] , 502 (1897). (7) Odet and Vignon, C. R. Acad. Sci., Paris, 69 [II], 1142 (1869). (8) F. Francis, B e r . , 39, 3798 (1906). (9) V. Gold, E. D. Hughes, and C. K. Ingold, J. Chem. Soc., 2467 (1850). (10) F. Daniels and E . H. Johnston, J . Amer. Chem. Soc., 43, 53 (1921); H. Eyring and F. Daniels, ibid., 52, 1472 (1930). (11) P. B. D. de la Mare. and J. H. Ridd, "Aromatic Substitution. Nitration, and Halogenation," Butterworth and Co. (Publishers) Ltd., London, 1959, p 5.1 ff. (12) V. Gold, E. D. Hughes, C. K. Ingold, and G. H. Williams, J . Chem. Soc., 2452 (1950). (13) J. J. Wren, J . Chromatogr., 4, 173 (lg60); G. J. Young, J . Colloid Sci., 13, 67 (1958). (14) K. B. Ebert, Monatsh. Chem., 88, 275 (1957).
neously by homogeneous and heterogeneous paths; AgI is a good surface catalyst and the reaction is autocatalytic.15 Inhibition of the CC14-AgNOa reaction by benzene is difficult to account for in homogeneous solution because the formation constant for C6H6Ag+ is small.le However, if Ag+ adsorbed on an initially small amount of AgCll' is the active reagent, this might be susceptible to more complete complexation by C6Ha. The comparatively poor performance of AgNOa-alumina may stem from a smaller surface area. We have given elsewhere some examples of nitrations by CC14-AgN03-silicic acid under chromatographic conditions.' Should this system ever be used to deliberately nitrate small samples, Iimitations of solubility in CCL would be encountered with many compounds, since most other solvents except saturated hydrocarbons react with the reagent. We investigated the possibility of nitrating 2,4-dinitrophenol, which is poorly soluble in carbon tetrachloride, by stirring a suspension with CCld-AgNO3-silicic acid. This produced a clean reaction mixture and 49% picric acid in 24 hr a t room temperature, along with 45% recovered starting material.
Experimental Section General.-All reactions were conducted at 24 f.2". Product analyses, except where otherwise stated, were via gas chromatography using silicone QF-1 columns (20oj, on acid-washed Chromosorb W ) in a F & M 700 instrument with a thermal conductivity detector. Peak areas from duplicate 2 0 4 samples were planimetrically compared with those from standard solutions prepared from Fisher Certified Reagent or Eastman White Label materials and chromatographed before and after the unknown. Linearity of detector response was verified over the concentration range employed. Nitrobenzene was chromatographed at 150°, the nitroanisoles at 200'. The analytical precision was -1%. Melting points were delermined by hot stage microscopy and are uncorrected. Mass spectra were measured on a GEC-AEI MS-12 instrument. Infrared and electronic spectra were recorded on Perkin-Elmer Model 337 and 202 instruments. Silver Nitrate Impregnated Silicic Acid and Aluminum Oxide. -3lallinckrodt chromatographic silicic acid (6.1 70water loss at 150°), 56.1 g, was slurried with 160 ml of methanol and added to a solution of 14.00 g of silver nitrate in 60 ml of 50% aqueous methanol. The resulting slurry was evaporated at 40-55" and 15 Torr in a rotary evaporator t o yield a white powder which was lightly ground in a Pyrex mortar. The product was dried to constant weight at 150" just before use, and the resulting samples thus contained 1.24 mmol of AgN03/g. Material which was dried to constant weight at 120" possessed the same chemical properties. Impregnated alumina was prepared in the same way, using Merck acid-washed chromatographic alumina and a nominal 15 wt yo loading rather than the above 207,; the dried product contained 0.932 mmol of AgNOs/g. Observations on CCl4-AgNOS-Silicic Acid Mixtures.-(a) A 50 mg sample of AgNOa-silicic acid was covered with CC1,; after 1 hr it was frozen, evacuated, and allowed to warm until a vapor pressure of ca. lo-* Torr resulted. The vapor was then observed mass spectrometrically, using an exciting potential of 70 eV. After subtracting the lines due to CCla, the following peaks remained: m/e (relative abundance) 36 (25), 38 (lo), 44 (92), 63 (loo), 64 (1.2), 65 (36), 98 (9.3), 100 (5.4), 102 (1.1). A similar experiment using CC14adsorbed on an excess of AgNOssilicic acid gave the following spectrum: 30 (2.6), 3.5 (4.9), 36 (0.5), 37 (1.4), 38 (O.l), 44 (loo), 45 (2.4), 46 ( l . l ) , 63 (18),
(15) J. M. Austin. 0. D . E.-% Ibrahim, and M. Spiro, J . Chem. S O C .B , 669 (1969). (16) L. J. Andrew and R. M. Keefer, "Molecular Complexes in Organic Chemistry," Holden-Day, Sa, Francisco, Calif., 1964, p 90 ff. (17) K. Fajans and K. yon Beckerath, Z . Physik. Chem., 97, 47 (1921).
SILVERNITRATEIMPREGNATED SILICICACID 65 (5.0), 69 (0.4), 70 (0.6), 72 (0.5), 98 (0.4), 100 (O.l), 121 (0.6), 131 (0.2). (b) A tightly stoppered sample of AgNO3-silicic acid covered with CCla was allowed to stand for 24 hr. A 1 ml sample of the liquid phase was transferred to a 10-cm path length fused-quartz cell. The uv-visible spectra were recorded using a slit-cell geometry which placed only the vapor above the CClr solution in the light path. The observed spectrum, consisting of cu. 40 narrow lines superimposed on broad maxima centered near 340 and 400 mp, was that reported for N0g-N2O4.l8 (c) A 3.868-g (4.80 mmol AgN03) sample of AgNOa-silicic acid was placed in a small chromatographic column arranged to deliver effluent into a solution containing 15 ml of CCld and 1.6 ml of aniline. Flow of carbon tetrachloride through the column was begun, and within 2 min a white precipitate appeared in the magnetically stirred receiver solution. Three fractions were collected using fresh aniline-CC14 solutions: effluent volumes in ml (flow times in min) 30 (20), 30 (35), and 65 (110). Each fraction was filtered, and the white solid was washed with water and dried, yielding 230, 34, and 15 mg of carbanilide respectively (total 279 mg, 1.32 mmol, 0.274 mmol COClz/mmol AgN03). The combined material was recrystallized (78% recovery) from ethanol, mp 243.5-244", mmp [with authentic carbanilide (mp 243.5-244')l 243-243.5'. If the elution is conducted with 10% of benzene in the cc14, very little phosgene is produced. (d) I n a similar experiment benzene replaced the aniline of experiment c. The effluent was stirred for 48 hr; analysis by glpc revealed no nitrobenzene. (e) A 2.336-g (2.89 mmol AgNOa) sample of AgNO3-silicic acid was covered with CClr in a vessel connected to a vacuum line by way of a small trap cooled in liquid Nz. Over a period of 2.5 hr, all of the volatile material was slowly distilled into the trap. The trapped material was warmed to --50" and 11 ml of a solution prepared by dissolving 5.07 g of NaOH in 50 ml of water was added. The mixture was allowed to come to room temperature, the phases were equilibrated and separated, and a water wash of the CCl4 phase was added to the aqueous phase which was then acidified with nitric acid. Gravimetric determination of C1- then produced 0.234 g (1.38 mmol) of silver chloride corresponding to 0.69 mmol of COClz or 0.24 mmol of COClg/mmol of AgNOs Nitration of Benzene (Expt 1-16) .-To the freshly dried AgNO3-silicic acid (or alumina) sample (weight in Table I ) was added 10 ml of CCld-hexane ( 1 : 2 ) . After 1-15 min, 250 ~1 (214 mg, 2.74 mmol) of benzene was injected and the mixture was agitated and allowed to stand for the specified time. It was then transferred to a small chromatography column and eluted with ether to provide one or two (depending on the amount of silicic acid used) 25.0-ml product solutions which were analyzed by gas chromatography.
.
(18) T. C. Hall, Jr., and F. E. Blacet, J . Chem. Phye., 20, 1745 (1952).
J. Org. Chem., Vol. 35, No. 8, 1970 2725 Nitration of Anisole (Expt 17-23).-Kitration with excess CCla (expt 17) was done as in the preceding paragraph, substituting anisole for benzene. Nitration with CC14 as limiting reagent (expt 19-23) was done by adding 10-20 ml of hexane to the dry, 5-10-g AgN03-silicic acid sample, injecting 100 pl (150 mg, 0.975 mmol) of cc14, and stirring magnetically for the specified time (Table I). Anisole (1.0 ml, 990 mg, 9.1 mmol) was injected and the mixture was stirred for 1 hr, then allowed to stand for 17 hr; work-up was as in the preceding paragraph. The 0- and p-nitroanisole peaks in the gas chromatogram were -75% resolved, and the isomer ratio was calculated by integration in comparison with standard mixtures. Nitration by HN03-Silicic Acid.-To a slurry of 3.18 g of dry (130-140' overnight) silicic acid in 10 ml of hexane-CCI4 (2: 1 ) was added 50 pl (73 mg, 1.28 mmol) of white, fuming nitric acid. The mixture was stirred magnetically for 5 min, and 125 pl (107 mg, 1.37 mmol) of benzene was added. The mixture was worked up and analyzed as in the reaction with CCl4 after 48 (96) hr to give 0.15 (0.17) mmol, 11 (137,), of nitrobenzene. Application of this procedure to 6.67 g of silicic acid, 16.5 ml of hexane, 73 mg (0.96 mmol) of nitric acid, and 1.0 ml (9.1 mmol) of anisole gave, after 18 hr, 0.29 mmol (317') of nitroanisoles. Nitration of 2,4-Dinitrophenol.-A suspension of 500 mg (2.72 mmol) of 2,4-dinitrophenol (mp 112.5-113.5') in 15 ml of ccl4 was vigorously stirred with 4.687 g (5.81 mmol AgNOa) of AgN03silicic acid for 24 hr. The suspension was filtered and the solid was washed with ether and methanol. The residue from evaporation of the filtrate was dissolved in 25 ml of methylene chloride and extracted with 180 ml of pH 4.4 citrate buffer. The buffer was washed with three 25-ml portions of methylene chloride and the combined organic phases were evaporated to 223 mg (1.21 mmol, 45y0) of recovered 2,4-dinitrophenol. The aqueous phase was acidified with 8 ml of sulfuric acid and extracted with four 28-ml portions of methylene chloride which gave on evaporation 306 mg (1.33 mmol, 4970) of picric acid, mp 120-122', mmp [with authentic material (mp 121-122.5')] 121- 122'; the infrared spectra were identical.
Registry No.-Benzene, 71-43-2; anisole, 100-66-3; silver nitrate, 7761-88-8; carbon tetrachloride, 56-230.
Acknowledgment. -This work was supported in part by a Kent State University Summer Faculty Research Fellowship. The writer thanks Mr. R. A. Champa for the mass spectra. Financial assistance of the National Science Foundation in acquiring the mass spectrometer is gratefully acknowledged.